Extending the history of child obesity in the United States: The Fels Longitudinal Study, birth years 1930-1993

Abstract

Objective: Little is known about the prevalence of child obesity in the US before the first national survey in 1963. There is disagreement about whether the obesity epidemic is entirely a recent phenomenon or a continuation of longstanding trends.

Design and methods: The BMIs of 1,116 children who participated in the Fels Longitudinal Study near Dayton, Ohio were analyzed. Children were born between 1930 and 1993 and measured between 3 and 18 years of age.

Results: Between the birth cohorts of 1930 and 1993, the prevalence of obesity rose from 0 to 14% among boys and from 2 to 12% among girls. The prevalence of overweight rose from 10 to 28% among boys and from 9 to 21% among girls. The mean BMI Z score rose from +0.25 to +0.72 among boys and from -0.11 to +0.26 among girls. Among boys, all these increases began after birth year 1970. Among girls, obesity began to rise after birth year 1980, but overweight and BMI Z-scores were already rising as early as the 1930s and 1940s.

Conclusions: Most of the results suggest that the child obesity epidemic was recent and sudden. The recency of the epidemic offers some hope that it may be reversed.

Figure 1

Smoothed trends of prevalences and mean BMI Z-score based on the extended IOTF standards. Solid lines are quadratic splines with a single knot in 1963; dashed lines are 95% pointwise confidence intervals; points are prevalences or means within birth year groupings corresponding to those in Table 1. P-values are for tests of the null hypothesis that the trend is flat. The trend in girls’ overweight was also fit to a straight line; under that simplified model, the trend was significant at P=.01.

Figure 2

Smoothed trends in prevalences and mean BMI z-score. The methods and data are the same as in Figure 1, but here the definitions of obesity, overweight, underweight, and Z-score are based on the CDC standards^, rather than the IOTF standard.